Skip to content

Security: sharkysan/formancy.ai-data

SECURITY.md

Security Policy

Reporting a vulnerability

Please report security vulnerabilities through GitHub Security Advisories ("Report a vulnerability" on the Security tab) rather than a public issue.

We aim to acknowledge a report within 3 working days and to provide a remediation timeline within 10 working days.

Please do not include a working exploit in the initial report; a description of the vulnerability class and the affected component is enough to begin triage.

Scope

Formancy Data connects to a customer's existing database and writes to it. That makes it a higher-risk component than a form renderer, and the areas we consider highest risk, and are most interested in reports about, follow from it:

  • SQL injection through identifiers — a schema, table or column name that reaches a query without having come from approved metadata and been quoted by the adapter — or through a value that was interpolated rather than bound.
  • Authorisation bypass on record read, lookup, create or update: a hidden or disabled control treated as a permission, a field accepted that the policy did not list, a tenant filter that a request could influence.
  • Cross-tenant reference through a lookup: a foreign-key value that names a record the actor may not see, accepted because the record exists.
  • Metadata disclosure: a discovery or error path that reveals schema, constraint or row information the connection's policy does not permit.
  • Secret exposure: a connection string or credential in a log line, an error message, an API response or a published bundle.
  • Lost update or silent overwrite: a write that proceeds without the expected version, or an ambiguous write replayed automatically.

Nothing in the packages fetches a URL, and no form document ever carries one; a report that found a way to make the module reach the network on an outsider's instruction is in scope and serious.

There aren't any published security advisories